Accumulating evidence suggests that Alzheimer's disease (AD) pathology, such as plaques and tangles, begins 10-20 years prior to the earliest signs and symptoms of cognitive decline. This period, during which pathology is developing but individuals remain cognitively normal, has been referred to as 'pre-clinical AD'. One pathological feature that appears to distinguish individuals with pre-clinical AD from those with very early AD (very mildly demented with plaques and tangles) is that the earliest clinical symptoms coincide with neuronal and synaptic loss and/or dysfunction in certain brain regions. Given this observation and that promising disease-modifying treatments are on the horizon, it will be critically important to have biomarkers that: 1) correlate with the presence of AD pathology in the brain regardless of clinical status;2) predict with high likelihood the development of cognitive decline in individuals who are still cognitively normal but developing AD pathology (antecedent biomarkers);and 3) differentiate those individuals with very mild or uncertain dementia (mild cognitive impairment) who are most likely to experience cognitive decline. Over the last five years, we have found that: 1) low CSF A(342 is very sensitive and specific for determining the presence or absence of amyloid in the brain as assessed by imaging with Pittsburgh Compound B (PIB), regardless of clinical status;2) ratios of CSF tau/A(342 and ptau181/A(342 are highly predictive of progression from CDR 0 to CDR >0.5 over an average 3-4 year period;and 3) new potential biomarkers for AD (ATIII, ZAG, CNDP1, ACT) can be identified in CSF by unbiased proteomic techniques. We hypothesize that an assessment of the CSFand plasma proteome, including markers such as A042, tau, ptau181, and other proteins, can be combined to develop an accurate determination of dementia risk in cognitively normal elderly individuals. To test this hypothesis, we propose the following aims: Aim 1:To determine the ability of CSF AB42, tau, ptaul81, ATIII, ACT, ZAG, and CNDP1, both alone and in combination, to predict clinical progression from CDR 0 to CDR >0 and progression from CDR 0.5 (very mild dementia) to CDR 1 (mild dementia);Aim 2: To determine whether CSF levels of Ap42, tau, ptaul81, ATIII, ACT, ZAG, and CNDP1, both alone and in combination, correlate with brain amyloid load as measured with PiB, brain volume as assessed by structural MRI, and neuropsychological test scores;Aim 3: To determine, in collaboration with the Wyss-Coray lab, whether a group of 18 plasma signaling proteins that was recently shown to classify AD vs. control subjects correlates with CSF biomarkers of AD studied in Aims 1 and 2, as well as whether these 18 proteins predict progression from CDR 0 to CDR >0 or from CDR 0.5 to CDR 1;and Aim 4: To identify novel CSF biomarkers of AD utilizing a new quantitative and highly sensitive proteomic technique called targeted label free LC-MS/MS analysis.